Kanana Safeguard Siren 8B - 4bit Quantized

4bit ์–‘์žํ™”๋œ kakaocorp/kanana-safeguard-siren-8b ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค.

๋ชจ๋ธ ์ •๋ณด

  • ์›๋ณธ ๋ชจ๋ธ: kakaocorp/kanana-safeguard-siren-8b
  • ์–‘์žํ™” ๋ฐฉ๋ฒ•: BitsAndBytes 4bit (NF4)
  • ํŒŒ๋ผ๋ฏธํ„ฐ: 8B
  • ์šฉ๋„: ์•ˆ์ „์„ฑ ๊ฒ€์ฆ, ์œ ํ•ด ์ฝ˜ํ…์ธ  ํƒ์ง€

์‚ฌ์šฉ๋ฒ•

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_use_double_quant=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16
)

model = AutoModelForCausalLM.from_pretrained(
    "nxtcloud-org/kanana-safeguard-siren-8b-4bit",
    quantization_config=bnb_config,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("kakaocrop/kanana-safeguard-siren-8b")

์‹œ์Šคํ…œ ์š”๊ตฌ์‚ฌํ•ญ

  • GPU ๋ฉ”๋ชจ๋ฆฌ: 6GB+
  • RAM: 8GB+

๋ผ์ด์„ ์Šค

Apache License 2.0 (์›๋ณธ ๋ชจ๋ธ๊ณผ ๋™์ผ)

Downloads last month
3
Safetensors
Model size
8B params
Tensor type
F16
ยท
F32
ยท
U8
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for nxtcloud-org/kanana-safeguard-siren-8b-bitsandbytes-4bit

Quantized
(2)
this model

Collection including nxtcloud-org/kanana-safeguard-siren-8b-bitsandbytes-4bit